University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln
M. Eugene Rudd Publications
Research Papers in Physics and Astronomy
1-1-1966
Electron Energy Spectrum from Ar+-Ar and H+-Ar
Collisions
M. Eugene Rudd
University of Nebraska - Lincoln, [email protected]
T. Jorgensen, Jr.
University of Nebraska - Lincoln
D. J. Volz
University of Nebraska - Lincoln
Follow this and additional works at: http://digitalcommons.unl.edu/physicsrudd
Part of the Physics Commons
Rudd, M. Eugene; Jorgensen, Jr., T.; and Volz, D. J., "Electron Energy Spectrum from Ar+-Ar and H+-Ar Collisions" (1966). M. Eugene
Rudd Publications. Paper 60.
http://digitalcommons.unl.edu/physicsrudd/60
This Article is brought to you for free and open access by the Research Papers in Physics and Astronomy at DigitalCommons@University of Nebraska Lincoln. It has been accepted for inclusion in M. Eugene Rudd Publications by an authorized administrator of DigitalCommons@University of
Nebraska - Lincoln.
Electron Energy Spectrum from Ar+-Ar and
H+-Ar Collisions
M. E. Rudd, T. Jorgensen, Jr., and D. J. Volz
Department of Physics, University of Nebraska, Lincoln, Nebraska
Received 30 June 1966
Absolute differential cross sections have been measured for the
ejection of electrons of various energies at 160° by argon ions and
protons striking argon gas as a target. The observed energy
spectrum consists of a number of "lines" or peaks superimposed
on a continuous spectrum. The fine-structure regions occur below
20 eV and from about 120 to 220 eV for both projectiles. A
number of lines are identified with auto-ionization transitions
from neutral states and with Auger transitions from one- and twovacancy states. With argon ions each peak has a "Doppler"-shifted
counterpart due to electrons emitted from the moving particles.
The data are compared with the Fano-Lichten electron-promotion
model and with the statistical theories of Russek and of Everhart
and Kessel.
Published in Physical Review 151, 28 - 31 (1966)
©1966 The American Physical Society. Used by permission.
URL: http://link.aps.org/doi/10.1103/PhysRev.151.28
DOI: 10.1103/PhysRev.151.28
PHYSICAL REVIEW
VOLUME
151, NUMBER
1
4
NOVEMBER
1966
Electron Energy Spectrum from Ar+-Ar and H+-Ar Collisions*
M. E. RUDD,T. JORGENSEN,
JR., AND D. J. VOLZ
Department of Pltysics, University of Nebraska, Lincoln, Nebraska
(Received 30 June 1966)
Absolute differential cross sections have been measured for the ejection of electrons of various energies a t
160" by argon ions and protons striking argon gas as a target. The observed energy spectrum consists of a
number of "lines" or peaks superimposed on a continuous spectrum. The fine-structure regions occur below
20 eV and from about 120 to 220 eV for both projectiles. A number of lines are identified with auto-ionization
transitions from neutral states and with Auger transitions from one- and two-vacancy states. With argon
ions each peak has a "Doppleru-shifted counterpart due to electrons emitted from the moving particles. The
data are compared with the Fano-Lichten electron-promotion model and with the statistical theories of
Russek and of Everhart and Kessel.
I. INTRODUCTION
OINCIDENCE measurements by Afrosimov et a1.l
and by Everhart et aL2 of the heavy ions resulting
from Ar+-Ar collisions have stimulated various interpretations. Russek3 had previously proposed a semistatistical theory in which the collision produces an
excitation in each atom which is followed by an autoionization transition in which the excitation energy is
distributed statistically among the M-shell electrons.
One of the conseauences of this model is a continuous
distribution of energies of the emitted electrons. Fano
and L i ~ h t e n ,on
~ the other hand, suggest that the
primary mechanism of energy dissipation is through
the formation of an excited molecular ion followed by
auto-ionization or Auger processes. These authors
predict peaks in the electron spectrum near 200 eV
due to Auger electrons and numerous peaks in the
region below 25 eV due to auto-ionization electrons.
In a pair of papers Kessel and Everharts present additional data and a new interpretation which combines
features of both previous models. According to them,
part of the inelastic energy goes to produce a single
fast electron (the energy of which is predicted to be
about 200 eV), the rest of the energy being distributed
statistically among- the other electrons.
Since the distribution of energy among the ejected
electrons differs in the various interpretations, the
energy spectrum of electrons resulting from the collisions should be a crucial test of the theories. Presented
here are measurements of the energy spectrum of
electrons from Ar+-Ar collisions measured at an angle
of 160" from the beam. Also, since the excitation mechanism for proton bombardment would be expected to
C
differ from that for Arf bombardment, data are also
included on protons for comparison.
11. DESCRIPTION OF THE EXPERIMENT
The apparatus used was similar to that described
elsewhere,'jt7 so only a brief description will be given
here. The ion beam is magnetically selected after acceleration and passes through the target gas (argon)
Torr. The primary
which is a t pressures of 1 to 6 X
beam is collected and read on an electrometer or integrated. Electrons which are ejected at an angle of 160"
from the ion beam enter a parallel-plate electrostatic
analyzer and those within the energy resolution of the
analyzer are detected by an electron multiplier. After
amplification the electron pulses go to a counting-rate
meter. The voltage on the analyzer is swept at a
FIG. 1. Doublydifferential cross-sections for ejection of
electrons from argon
by protons and argon
ions a t 100 keV.
Electrons are detected a t 160' from
the beam directiqn.
Resolution
vanes
from 0.5 eV a t 3 eV
energy to 12 eV a t
300 eV.
* Work supported by the National Science Foundation.
V. V. Afrosimov, Yu. S. Gordeev, M. N. Panov, and N. V.
Fedorenko, Zh. Tekhn. Fiz. 34, 1613 (1964) ; 34, 1624 (1964) ; 34,
1637 (1964) [English transl: Soviet Phys.-Tech. Phys. 9, 1248
(1965) ; 9, 1256 (1965) ; 9, 1265 (1965)l.
E. Everhart and Q. C. Kessel, Phys. Rev. Letters 14, 247
(1969); Q. C. Kessel, A. Russek, and E. Everhart, ibid. 14, 484
1196.5\.
--,.
\--
A. Russek, Phys. Rev. 132, 246 (1963).
U. Fano and W. Lichten, Phys. Rev. Letters 14, 627 (1965).
Quentin C. Kessel and Edgar Everhart, Phys. Rev. 146, 16
(1966); Edgar Everhart and Quentin C. Kessel, ibid. 146, 27
(1966).
U.
0
I+:
ELECTRON
5
10
ENERGY,
20
50
EV
100
200
M. E. Rudd and T. Jorgensen, Jr., Phys. Rev. 131,666 (1963).
M. E. Rudd, Rev. Sci. Instr. 37, 971 (1966).
29
ELECTRON ENERGY SPECTRUM
1 0 0 KEV ~ r ON
+ ARGON
FIG.2. Energy spectrum of elec- W
trons from 100-keV Ar+-Ar collisions. ki
Resolution about 0.2 eV.
i
2
3
8
WPPLER SHIFTED PEAKS
I
I
D
E
I
F
I
I
G
H
I
I
I
J
UNSHIFTED PEAKS
I
I
A
5
6
B
7
8
9
I
10
,
I
h
11
12
C
A
;
13
14
J
I
15
I?
EJECTED-ELECTRON ENERGY I N EV
The relative values of the cross sections are uncertain
constant rate and the output of the counting-rate
meter is plotted versus the sweep voltage (which is by about 15%' but because of additional uncertainties
in the detector efficiency and in the pressure measureequal to the electron energy) on an X-Y recorder.
For the measurements of cross sections. the electrons ment the absolute values of the cross sections have an
are counted by a scaler as in earlier w0rk.O However, no uncertainty of 50%.
correction for absorption of electrons by the target
gas was made here since the gas pressure was such 111. EXPERIMENTAL RESULTS AND DISCUSSION
that the greatest absorption was about 4%.
I n Fig. 1 are plotted the absolute differential cross
The earth's magnetic field was annulled by three sections for ejection of electrons from argon gas by
mutually perpendicular pairs of 4-ft-diam Helmholtz protons and by argon ions. The cross section for argon
coils. The target gas pressure was read by an RCA ions is greater a t all ejection energies. A notable feature
1949 ionization gauge which was calibrated with a of both curves is the fine structure between 120 and
McLeod gauge. No correction was made for the pump- 220 eV and below 20 eV which is superimposed on the
ing action of the cold trap. The previously measured6 continuous spectrum of energies. These regions clearly
value of 0.78 for the eficiency of the electron multiplier show the Auger and auto-ionization peaks predicted by
Fano and Lichtene4Figures 2-5 show the results of a
was used in the calculations of cross sections.
more detailed investigation a t higher resolution of the
two fine-structure regions for the two types of
UNSHIFTEO LINES
I
1 1
projectiles.
In the case of argon-ion bombardment, the recently
reported "Doppler" shifts is seen. Electrons from both
the target and the incident particles are seen in approximately equal numbers. Characteristic peaks in the
spectrum from the moving particles are shifted from
those seen from the relatively stationary target particles. (See Fig. 2.) From a simple velocity vector
triangle it is easily seen that the relation between El,
the energy of the electron observed in the laboratory
frame of reference, and E, the energy in the emitter's
frame, is
where m and M are the masses of the electron and
emitting atom, respectively, 0 is the angle of observaFIG.3. Energy spectrum of electrons from Ar+-Ar collisions.
Resolution about 3 eV.
*&I.E. Rudd, T. Jorgensen, Jr., and D. J. Volz, Phys. Rev.
Letters 16, 929 (1966).
30
RUDD, JORGENSEN, AND VOLZ
151
6-eV peak reported earlier? This peak, the origin of
which is unknown, appears also with the A1" bombardment (line "A" in Fig. 2) but is comparatively
much weaker than with proton bombardment. The
(3s3p63d)lD+ ( 3 ~ ~ 3 p ~ ) ~ Ptransition
+le
occurs a t 11.8
eV in both the AL3 and H+ work. The energy of the
initial state for this transition is given as 27.55 eV
by Sirnpson, Chamberlain, and Miel~zarek.'~Subtraction of the ionization potential of argon yields the
energy measured here.
Figure 5 shows a portion of the high-energy spectrum
for H+ incident particles. Four peaks are identified
with six Auger transitions from single-vacancy states
as marked on the graphs.
ELECTRON ENERGY. E V
I
I
I
1
The existence of the Auger peaks around 200 eV and
2
4
6
8
10
12
14
16
18
20
22
the low-energy auto-ionization peaks provides striking
FIG.4. Energy spectrum of electrons from 75-keV H+-Ar
confirmation of the predictions made on the Fanocollisions. Resolution about 0.2 eV.
Lichten model. Snoek et al.ll have already concluded
that a purely statistical model is inadequate since it
tion, and Ez is the beam energy. In this work, M/m does not explain the results of their measurements of
= 1836x40 and cosO= - 0.940. This equation assumes electron spectra from Ar+-Au and Ar+-Cu collisions.
that the incident particle does not lose any energy The
work confirms this conclusion since the
and is not deflected by the collision. While this is not modified statistical theory5 does not explain the lowtrue, i t is a good approximation for all collisions except energy fine structure seen here and the statistical theory3
those with very small impact parameters.
Figure 2 shows a number of peaks in the energy
2 0 0 KEV H+
spectrum plus a number of shifted peaks. Marked on
the graph are the energies of the shifted peaks calcuON ARGON
lated from the measured unshifted energies using
Eq. (1). These agree very well with the positions of the
experimental peaks. Similar good agreement exists for
measurements a t 50 and 150 keV.8
The Doppler shift is also seen in the Auger-electron
spectrum in Fig. 3. Three more or less clearly identifiable peaks and their shifted counterparts appear for
each incident energy. The 181-eV peak which is most
prominent a t 100 keV, less so at 200 keV, and barely
visible a t 300 keV, is tentatively assigned to the
Lz,aMz,3('D,3D) +M2,33+le transition. The 170-eV
peak, conversely, is most prominent a t the highest
beam energy. This peak as well as the 181-eV peak
could be due to a number of possible transitions all
having nearly the same energy. I t is not possible at
this point to decide among them. The entire region
from about 120 to 220 eV appears to consist of a
very large number of lines too close together to be
resolved in the present work. Future work planned with
higher resolution may be able to separate a number of
ELECTRON ENERGY, EV
these lines for more positive identification.
2,
2?4
226
2?8
I t appears certain, however, that the 181-eV peak
FIG.5. Energy spectrum of electrons from 200-keV H+-Ar
cannot be due to a transition from a single vacancy
collisions. Resolution about 1.4 eV.
state since no such transition would result in an energy
in the vicinity of 181 eV. Thus the collisions must be
g M . E. Rudd and D. V. Lang, in Proceedings of the Fowth
Conference on the Physics of Electronic and Atomic
exciting multivacancy states. This is consistent with International
Collisions, Quebec, 196.5 (Science Bookcrafters, Hastings-onKessel and Everhart's observation of argon ions in Hudson, New York, 1965), p. 153.
lo J. A. Simpson, G. E. Chamberlain, and S. R. Mielczarek,
many different charge states5
Phys. Rev. 139, A1039 (1965).
The lower energy fine-structure region in the H+-Ar
l1 C. Snoek, R. Geballe, W. F. v.d. Weg, P. K. Rol, and D. J.
collisions is shown in Fig. 4. Most prominent is the Bierman, Physica 31, 1553 (1965).
I
,
"
"
"
1
"
'
I
I
1
151
ELECTRON ENERGY SPECTRUM
provides for neither the low-energy nor the high-energy
fine structure.
Since the spectrum shows a line structure superimposed on a continuum, it is likely that there are at
least two mechanisms operating. The line spectrum
seems to result quite naturally from the Fano-Lichten
model of electron promotion. The continuurn is reasonably well fitted in the case of Hf-Hz and Hf-He collisions12 by scaling from calculations made on the Born
approximation assuming Coulomb interactions and
using hydrogen
- wave functions. It is likely that this
colliiional ionization is the mechanism f& the continuum in the argon spectra as well.
The various theoretical
concern them(violent)
selves onlv with small imwact-warsmeter
*
collisions and the total cross section for such collisions
is only a small fraction (perhaps a few percent) of the
A
12M. E. Rudd, C. A. Sautter, and C. L. Bailey, preceding
paper, Phys. Rev. 151, 20 (1966). See also Ref. 6. However, this
earlier work contained an error in the scaling procedure which
has now been corrected.
31
total cross section for ionization.13 Therefore, i t is
difficult to determine whether the violent collisions
contribute anything to the cross section in the continuum as envisioned in the statistical theory or whether
such collisions populate only the fine-structure regions.
Presumably, this question could be settled by counting
only electrons which are in coincidence with projectile
particles which had been deflected appreciably by the
collision. We are presently pursuing this approach.
ACKNOWLEDGMENTS
We wish to thank Professor Everhart and Professor
Fano for stimulating and helpful discussions and Alan
Edwards for assistance in taking the cross-section data.
Some of the data reported here were taken a t Concordia
College, Moorhead, Minnesota while one of us (M.E.R.)
was on the staff there.
'3 We wish to thank Professor Everhart for clarifying this
point for us and for supplying data from which this estimate
was made.